JP3223339B2 - controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an IMC (Internal Model)
The present invention relates to a controller using a control algorithm having a control structure, and more particularly to a controller that can perform good control even when adjustment of the time constant of a filter cannot be performed.

[0002]

2. Description of the Related Art Conventionally, there has been proposed a controller using a control algorithm having an IMC structure for performing control by incorporating an internal model expressing a process to be controlled mathematically.
If this IMC controller is used, a large dead time (from the time when the hot air comes out of the air conditioner to the time when the indoor temperature rises) is increased in the process to be controlled (for example, when the controller is used for an indoor air conditioner, this corresponds to the indoor environment) There is an excellent advantage that it is possible to cope even if there is (time).

FIG. 17 is a block diagram of a control system using a conventional IMC controller. Reference numeral 33 denotes a first subtraction processing unit for subtracting a feedback amount to be described later from the target value;
Reference numeral 2 denotes a filter unit for preventing a change in the output of the first subtraction processing unit 33 from being transmitted suddenly, and 34 denotes a filter unit 3
An operation unit 36 for calculating an operation amount which is an output of the controller based on the output of 2, an internal unit 36 which is obtained by approximating a process to be controlled by a mathematical expression and outputs a reference control amount corresponding to a control amount which is a control result; The model 38 is a second subtraction processing unit that subtracts the reference control amount from the internal model 36 from the control amount and outputs a feedback amount, and 40 is a control target process.

Further, F, Gc, Gm, and Gp are transfer functions of the filter unit 32, the operation unit 34, the internal model 36, and the controlled process 40, r is a target value, u is an operation amount, and y
Is a control amount, ym is a reference control amount, and e is a feedback amount.

Next, the operation of such an IMC controller will be described. First, the feedback amount e is subtracted from the target value r by the first subtraction processing unit 33, the operation amount u is calculated from the output of the filter unit 32 by the operation unit 34, and the operation amount u is calculated by the control unit process 40 and the internal model 36. Is output. Then, the reference control amount ym from the internal model 36 is subtracted from the control amount y of the process 40 by the second subtraction processing unit 38, and the result is fed back to the first subtraction processing unit 33 as the feedback amount e. A control system is configured.

The parameters of the internal model 36 of the IMC controller are set such that the transfer function of the internal model 36 and the transfer function (approximate expression) of the process 40 match based on the result of identifying the process 40 to be controlled. Is done. Further, the time constant of the filter unit 32 is determined in consideration of the balance between the control responsiveness and the stability. When the responsiveness is emphasized, the time constant of the filter is set to a smaller value, and when the stability is emphasized, the time constant is set to a larger value.

[0007]

The conventional IMC controller is configured as described above, and when the specification relating to the control responsiveness and stability is changed, the adjustment is performed by changing the time constant of the filter. However, the amount of operation for process control has physical upper and lower limits,
Even if the responsiveness of the filter is adjusted in order to improve the responsiveness of the target value tracking, there is a problem that the operation amount reaches a plateau due to the upper and lower limiters and the effect of the responsiveness improvement may not be obtained.

In addition, when the process to be controlled has a large variation in the dead time, the control is stabilized by adjusting the stability of the filter. However, if the dead time of the process becomes 3 to 4 times or more as long as the dead time of the internal model, the filter time constant must be adjusted at the expense of responsiveness, and the IMC controller is effective for processes with large dead time. There is a problem that the advantage of the above cannot be obtained. The present invention has been made to solve the above-described problem, and has as its object to provide a controller having an IMC structure that can perform good control even when adjustment of the filter time constant cannot be performed.

[0009]

SUMMARY OF THE INVENTION According to the present invention, a target value filter section for outputting a target value of an input control with a transfer function having a time delay characteristic, and a feedback amount is subtracted from the output of the target value filter section. A first subtraction processing unit,
A target value / disturbance filter that outputs the output of the subtraction processing unit with a time delay characteristic, and the transfer function of the internal model excluding the dead time element based on the parameters of the internal model.
An operation amount calculation unit including an operation unit that calculates and outputs an operation amount from the output of the target value / disturbance filter unit based on the characteristic of the reciprocal of the number, an internal model storage unit that stores internal model parameters, and The internal model
An internal model output operation unit for calculating the reference control amount from the operation amount according to the characteristic of the attainment function, and a feedback amount by subtracting the reference control amount output from the internal model output operation unit from the control amount of the control target process. A subtraction processing unit of 2;
An internal model parameter is calculated based on a process parameter obtained by model identification of a process to be controlled and an adjustment coefficient for designating a control characteristic emphasizing responsiveness or stability, and stored in an internal model storage unit. And a model parameter calculation unit.

In addition, the transfer function of the internal model has a gain and
And internal model parameters including time constants.
When the adjustment coefficient indicates the importance of responsiveness , the internal model parameter calculation unit determines the value obtained by multiplying the gain and the time constant in the process parameter by a specific value of 1 or less as the gain and the time constant in the internal model parameter and adjusts the adjustment. When the coefficient indicates importance on stability, a value obtained by multiplying the gain and the time constant in the process parameter by a specific value larger than 1 is used as the gain and the time constant in the internal model parameter.

[0011]

[0012]

According to the present invention, a feedback control system is constituted by the target value filter section, the first subtraction processing section, the manipulated variable operation section, the internal model storage section, the internal model output operation section, and the second subtraction processing section. Have been. Then, the setting of the internal model is such that the internal model parameter calculation unit calculates the internal model parameter from the process parameter and the adjustment coefficient,
This is performed by storing this in the internal model storage unit. Also, the internal model parameter calculation unit multiplies the gain and time constant in the process parameter by a specific value of 1 or less when the adjustment coefficient indicates responsiveness emphasis, and specifies the value greater than 1 when the adjustment coefficient indicates stability emphasis. The gain and the time constant in the internal model parameters are calculated by multiplying the values.

[0013]

[0014]

FIG. 1 is a block diagram of a controller having an IMC structure showing one embodiment of the present invention, FIG. 2 is a block diagram of a control system using the controller, and FIG. 3 is a diagram for explaining the operation of the controller. It is a flowchart figure.

In FIG. 1, reference numeral 1 denotes a target value input unit for inputting a target value r set by an operator to a controller, and 2 denotes a characteristic in which the transfer function of the target value r from the target value input unit 1 is a first-order lag. The target value filter section 3 outputs the feedback amount e from the output rf of the target value filter section 2.
Is a first subtraction processing unit, and 4 is an operation amount calculation unit that calculates an operation amount u from an output e1 of the first subtraction processing unit 3 based on parameters of an internal model storage unit described later. This is a signal output unit that outputs the operation amount u output from the unit 4 to a control target process not shown in FIG.

Reference numeral 6a denotes an internal model storage unit for storing the internal model parameters of the controller, and 6b an internal model output operation for performing an operation as an internal model based on the parameters of the internal model storage unit 6a and outputting a reference control amount ym. And 7, a control amount input unit for inputting a control amount y from the process to the controller, and 8 a reference control amount output from the internal model output operation unit 6b based on the control amount y output from the control amount input unit 7. This is a second subtraction processing unit that subtracts ym and outputs a feedback amount e.

Reference numeral 9 denotes an adjustment coefficient input unit for inputting an adjustment coefficient, which indicates an operator's required specification regarding the balance between control stability and responsiveness, to the controller. Reference numeral 10 denotes a model identification of a process to be controlled. A process parameter input unit 11 for inputting process parameters to the controller is an internal model parameter calculation unit that calculates an internal model parameter based on the adjustment coefficient and the process parameter and stores the calculated internal model parameter in the internal model storage unit 6a.

In FIG. 2, reference numeral 4a denotes a target value / disturbance filter unit which is inside the manipulated variable operation unit 4 and outputs the output e1 of the first subtraction processing unit 3 with a transfer function having a first-order delay characteristic.
4b is an operation unit for calculating an operation amount u from the output of the target value / disturbance filter unit 4a, 6 is an internal model comprising an internal model storage unit 6a and an internal model output operation unit 6b, and F1 is a target value. Transfer function of filter unit 2, F
Reference numeral 2 denotes a transfer function of the target value / disturbance filter unit 4a.

FIG. 2 shows the target value filter unit 2 of FIG.
The basic configuration of the controller including the first subtraction processing unit 3, the operation amount calculation unit 4, the internal model storage unit 6a, the internal model output calculation unit 6b, and the second subtraction processing unit 8 includes the control target process 40. It has been rewritten as a control system.

In the controller according to the present embodiment, the internal model parameter calculation unit 11 calculates the parameters of the internal model 6, whereby the characteristics of the target value filter unit 2, the operation amount calculation unit 4, and the internal model output calculation unit 6b are changed. Although the operation of the control system is determined, the operation of the controller as the control system will be described first.

The target value r is set by the operator of the controller, and is input to the target value filter unit 2 via the target value input unit 1 (step 104 in FIG. 3). The target value filter unit 2 calculates the output value rf from the input target value r as in the following equation (step 105). rf = F1 × r (1) F1 = 1 / (1 + T1 × s) (2)

The time constant T1 in the transfer function F1 in the equation (2) is set as follows according to the time constant Tm of the internal model 6 described later. T1 = 4 × α × Tm (3) α is a proportionality constant, for example, α = 0.168. Next, the first subtraction processing unit 3 subtracts the feedback amount e2 output from the second subtraction processing unit 8 from rf-e2, that is, the output rf of the target value filter unit 2 (step 106).

The target value / disturbance filter section 4a in the manipulated variable operation section 4 outputs the output e1 of the first subtraction processing section 3 with a transfer function F2 having the time constant T2 as shown in the following equation. F2 = 1 / (1 + T2 × s) (4) The time constant T2 is the same as the time constant T1 in the internal model 6.
The time constant Tm is set as follows. T2 = α × Tm (5)

The operation unit 4 in the operation amount calculation unit 4
b is the manipulated variable u from the output of the target value / disturbance filter unit 4a.
Is calculated, and its transfer function Gc is stored in the internal model storage unit 6
The following equation is obtained based on the gain and the time constant of the internal model 6 output from a, and is the reciprocal of the transfer function Gm of the internal model 6 excluding the element of the dead time Lm. Gc = (1 + Tm × s) / Km (6) where Km and Tm are gains of the internal model 6, respectively.
It is a time constant.

Therefore, the manipulated variable calculating section 4 calculates the manipulated variable u from the output e1 of the first subtraction processing section 3 as follows:
This is output to the control target process 40 (actually, an operating device such as a valve) via the signal output unit 5 and is output to the internal model output calculation unit 6b (step 107). u = F2 × Gc × e1 = [(1 + Tm × s) / {Km × (1 + T2 × s)}] × e1 (7)

Next, the control target process 40 is assumed to have elements of the first-order delay and the dead time, and its transfer function Gp
Can be represented by an approximate transfer function such as Gp = Kp × exp (−Lp × s) / (1 + Tp × s) (8) where Kp, Lp, and Tp are the gain, dead time, and time constant of the process 40, respectively.

In the internal model, the parameters Kp, Tp, and Lp obtained as a result of model identification of the process 40 in the example of FIG. 17 are directly used as the internal model gain Km, time constant Tm, and dead time Lm. The process 40 is represented by an approximate expression similar to the expression (8), but the internal model 6 of the present embodiment does not use the result of model identification as it is.

That is, the internal model output operation unit 6b
The reference control amount ym is calculated from the operation amount u based on the gain Km, the time constant Tm, and the dead time Lm stored in the internal model storage unit 6a as in the following equation (step 108).
These internal model parameters are calculated by the internal model parameter calculation unit 11 from the process parameters as described later. ym = Gm × u = {Km × exp (−Lm × s) / (1 + Tm × s)} × u (9)

Next, the control amount y is input to the control amount input unit 7 from the control target process 40 (actually, a sensor for detecting the control amount y) (step 109). Then, the second subtraction processing unit 8 subtracts the reference control amount ym from the internal model output calculation unit 6b from y-ym, that is, the control amount y, and outputs the subtraction result as a feedback amount e2 (step 110). . Step 104 to above
The operation of step 110 is repeated every control cycle until the controller is stopped by an instruction from the operator or the like (step 111).

This is the operation of the feedback control system which is the basic configuration of the controller having the IMC structure. Next, operations of the adjustment coefficient input unit 9, the process parameter input unit 10, and the internal model parameter calculation unit 11 that determine the above control characteristics will be described.

In the adjustment coefficient input section 9, an operator sets an adjustment coefficient C indicating the required specification of the operator regarding the balance between control stability and responsiveness (step 100 in FIG. 3). The operator sets the adjustment coefficient C to 0.1 ≦ C ≦ 1 when importance is placed on the responsiveness of control, and sets C> 1 when importance is placed on stability.

Next, the process parameter input unit 10 inputs the gain Kp of the process 40 obtained by using a conventional IMC parameter setting method for approximating the process 40 to be controlled as in equation (8) and identifying the model. , The time constant Tp,
The dead time Lp is set by the operator (step 101).

Then, the internal model parameter calculator 11
Calculates the gain Km, the time constant Tm, and the dead time Lm of the internal model 6 based on the adjustment coefficient C output from the adjustment coefficient input unit 9 and the process parameter output from the process parameter input unit 10 as follows: It is calculated (step 102). Km = C × Kp (10) Tm = C × Tp (11) Lm = Lp (12)

The gain Km, time constant Tm, and dead time Lm of the internal model 6 thus calculated are stored in the internal model storage unit 6.
The internal model 6 is set by being output to and stored in “a” (step 103). This means that the characteristics of the target value filter unit 2, the manipulated variable operation unit 4, and the internal model output operation unit 6b are determined. As a result, the characteristics of the control system are determined. Be started.

FIG. 4 is a diagram showing the target value follow-up performance when the conventional IMC controller is used for controlling the liquid level in the tank, and FIG. 5 similarly shows the target value follow-up performance of the controller of this embodiment. FIG. 4 and 5 show the target value r at 20 seconds.
Is changed from 0 to 4 cm, and a control result y (liquid level) of the control result is obtained as a simulation result. Here, the gain K of the process to be controlled, which is the liquid in the tank,
Let p be 1, the time constant Tp be 120 seconds, and the dead time Lp be 20 seconds.

In the conventional controller, the process parameters of the model identification result are stored in the internal model storage unit 6a without being calculated by the internal model parameter calculation unit 11 in the controller of the present embodiment. . Therefore, assuming that the model identification is correctly performed, the gain Km of the internal model of the controller is 1, the time constant Tm is 120 seconds, the dead time Lm is 20 seconds, and the time constant T1 of the target value filter is 80. 64
Second, the time constant T2 of the target value / disturbance filter section is 20.16
Seconds.

In FIG. 4, y1 is a control amount when these are used as they are, and y2 is a time constant of the filter, T1 = 2.
This is a control amount when the responsiveness adjustment is performed by changing to 0.16 seconds and T2 = 5.04 seconds. In the controller of the present embodiment, the adjustment coefficient C = 0.5 is set in the adjustment coefficient input unit 9 and the same process parameters as those of the conventional controller are set in the process parameter input unit 10, so that the internal model parameter calculation is performed. The unit 11 calculates the gain Km to be 0.5, the time constant Tm to be 60 seconds, and the dead time Lm to be 20 seconds, whereby the time constant T1 is 40.32 seconds and the time constant T2 is 10.08 seconds.

In the conventional controller shown in FIG. 4, the settling time when the filter adjustment is not performed (y1) is 281 seconds (where the settling time is a deviation range within 5% of the variation of the target value r, ie, 3. 8 <y <4.2), and when the responsiveness of the filter is adjusted (y2), 2
28 seconds. On the other hand, in the controller of this embodiment shown in FIG. 5, the time is set at 97 seconds, which is less than half, and it can be seen that the responsiveness is greatly improved.

The effect of improving the responsiveness is obtained for the following reasons. FIG. 6 is a diagram showing a change in the operation amount in the simulations of FIGS. 4 and 5, where u1 is the operation amount of the conventional controller, and u2 is the operation amount of this embodiment. However, the conventional controller shows only an example in which the responsiveness of the filter is adjusted. The manipulated variable u1 of the conventional controller rapidly rises to 170% at the maximum and then falls shortly thereafter. That is, it can be seen that the responsiveness adjustment of the filter is intended to improve the responsiveness by setting the operation amount to a very high value.

On the other hand, the operation amount u of the controller of this embodiment is
2 instead of rising only up to 110%,
Maintain a value close to 00% for a long time. In other words, it is understood that the responsiveness is to be improved by maintaining a higher operation amount for a long time. Therefore, if the operation amount u changes as shown in FIG. 6, the effect of improving the responsiveness should be obtained even with the conventional controller. However, the operation amount u is normalized to 0 to 100% due to physical restrictions (for example, the valve is broken when the valve is suddenly operated), and is not shown in the present embodiment and the conventional controller. The upper and lower limiters are set for the operation amount by the limiter means.

FIG. 7 is a diagram showing such an actual operation amount change. In the conventional controller, a high value of the operation amount reaches a peak, and the operation amount cannot be maintained for a time sufficient to compensate for the peak, so that the effect of improving the responsiveness cannot be obtained. On the other hand, in the controller of the present embodiment, even if the operation amount reaches a plateau due to the upper limit of the operation amount, the higher value is maintained for a long time, and as a result, the effect of improving the responsiveness can be obtained.

FIG. 8 is a view showing a target value follow-up characteristic when the conventional IMC controller is used for controlling the level of the liquid level in the tank similarly to the example of FIG. 4, and FIG. 9 is likewise an example of the present embodiment. It is a figure which shows the target value followability of a controller, and y3 is a control amount at the time of performing stability adjustment of a filter with a conventional controller. The process to be controlled here is the gain Kp
Is 4, the time constant Tp is 480 seconds, and the dead time Lp greatly varies up to 80 seconds at the maximum.

In the conventional controller, the parameters of the internal model are set to be the same as those in the example of FIG. 4 based on the result of model identification. Therefore, a model identification error of Kp / Km = Tp / Tm = 4 occurs, and the dead time Lm is also reduced. That is, an error of up to four times occurs. The time constant T1 of the target value filter unit and the time constant T2 of the target value / disturbance filter unit are the same as in the example of FIG. 4 when the filter adjustment is not performed (y1). T1 = 20 in y3)
1.6 seconds, T2 = 50.4 seconds.

Further, in the controller of this embodiment, the adjustment coefficient C = 2 is set in the adjustment coefficient input unit 9 and the same process parameters as those of the conventional controller are set in the process parameter input unit 10, so that the internal model parameters are set. The calculation unit 11 sets the gain Km to 2 and the time constant Tm to 24
0 seconds and the dead time Lm are calculated as 20 seconds, whereby the time constants T1 = 161.3 seconds and T2 = 40.32 seconds.

As is apparent from FIG. 8, in the conventional controller, when an identification error occurs, the oscillation of the control amount y cannot be suppressed even when the stability of the filter is adjusted (y3). It can be seen that the controller of the example is settled in the settling time of 1572 seconds, and the stability is improved, despite the same process parameter setting error as the conventional example.

In other words, when the identification of the process to be controlled includes an error or when the characteristics of the process fluctuate, it is necessary to set the control characteristics with an emphasis on stability. If not, it is advisable to increase the responsiveness of the control in order to improve the production efficiency. As described above, according to the controller of the present embodiment, it is possible to realize a control with good stabilization without sacrificing the responsiveness only by specifying the adjustment coefficient C emphasizing stability. Only by designating the adjustment coefficient C that is important, it is possible to realize control with a faster response than a normal IMC controller.

Incidentally, it is considered that the dead time fluctuation of the process to be controlled is particularly difficult to deal with as a factor of deteriorating the control characteristics. If the dead time fluctuation is about two to three times, the conventional filter time constant The adjustment or the controller shown in FIG. 1 can cope with the problem. However, if a further change in the dead time is expected, it cannot be coped with without sacrificing the responsiveness. Therefore, in such a case, FIG.
A different process from that of the example is required.

FIG. 10 is a block diagram of a controller having an IMC structure showing a reference example of the present invention, and FIG. 11 is a flowchart for explaining the operation of the controller. The same parts as in FIGS. The code is attached.

Reference numeral 11a denotes an internal model parameter calculation unit for performing the same parameter calculation as the internal model parameter calculation unit 11 of FIG. 1, and 12 denotes an internal model parameter calculation unit 11 based on an adjustment coefficient C and a mode signal indicating an operation mode.
a is a mode switching unit that switches the operation of a and an internal model parameter determination unit described later, and 13 is an internal model parameter determination unit that calculates internal model parameters that can respond to dead time fluctuations of the process to be controlled.

Also in this embodiment , the target value input unit 1, the target value filter unit 2, the first subtraction processing unit 3, the operation amount calculation unit 4, the signal output unit 5, the internal model storage unit 6a, and the internal model output calculation The operation of the basic configuration of the controller including the unit 6b, the control amount input unit 7, and the second subtraction processing unit 8 is exactly the same as the example in FIG. 1 (steps 104 to 111 in FIG. 11).
The operation of the adjustment coefficient input unit 9 (step 200) and the operation of the process parameter input unit 10 (step 201) are exactly the same.

A mode signal Q indicating an operation mode is set in the mode switching section 12 by an operator (step 202). The operator normally sets a mode signal Q = 0 indicating the first mode in the mode switching unit 12, and when it is expected that the process to be controlled has a large dead time variation, the mode signal Q = 2 indicating the second mode. Set 1. Next, the mode switching unit 12 determines the state of the set mode signal (step 203). If the mode signal is Q = 0, the mode switching unit 12 outputs this as it is and calculates the internal model parameters to the internal model parameter calculation unit 11a. Let it.

The internal model parameter calculating section 11a operates when the mode signal is Q = 0, and its operation is the same as that of the internal model parameter calculating section 11 in FIG. That is, the internal model parameter calculation unit 11a calculates the internal model parameters using the equations (10) to (12) (Step 204). Then, the internal model 6 is set by outputting and storing the calculated parameters to the internal model storage unit 6a (step 20).
5). Subsequent operations are exactly the same as in the example of FIG. 1 (steps 104 to 111).

Also, the mode switching unit 12 determines in step 20
If the mode signal is Q = 1 in step 3, the state of the adjustment coefficient C set in the adjustment coefficient input section 9 is determined (step 206). If so, the mode signal Q = 1 is output as it is, and the internal model parameter determination unit 13 calculates the internal model parameters.

The internal model parameter determination section 13 operates when the mode signal is Q = 1, and outputs the adjustment coefficient C output from the adjustment coefficient input section 9 and the process parameter output from the process parameter input section 10. The gain Km, the time constant Tm, and the dead time Lm of the internal model 6 are calculated based on the following equation (step 207). Km = C × Kp (13) Tm = Tp (14) Lm = Lp (15)

The operations after step 205 are the same as described above. Also, the mode switching unit 12 performs step 2
If the adjustment coefficient C is set to 0.1 ≦ C ≦ 1 and the responsiveness is emphasized in 06, the mode signal Q = 1 is changed to Q = 0 and output, and the parameter is sent to the internal model parameter calculation unit 11a. The calculation is performed (step 204).

Such a change is performed by the internal model parameter determination unit 13 to cope with the time delay of the process. If the time delay is large, the control becomes unstable. The reason is that the control characteristic that emphasizes the responsiveness should be selected, and the reason why the adjustment coefficient C that emphasizes the responsiveness is set when the mode signal is Q = 1 is considered to be an operator error. Therefore, such an operation of the mode switching unit 12 can cope with an operator's setting error.

FIG. 12 shows a conventional IMC controller shown in FIG.
Examples and shows a target value follow-up property when used to control the height of the liquid level in the same way in the tank, FIG. 13 is a diagram showing a target value follow-up of the controller of the present reference example as well, y4 Is a control amount by the controller in the example of FIG. Here, gain Kp, time constant Tp, dead time L of the process to be controlled
It is assumed that p is the same as in the example of FIG. 4 and that the same parameter setting as that of the example of FIG. 8 is performed for the conventional controller. And the same as

It is also assumed that the same parameter settings as in the example of FIG. 9 have been made for the controller of the example of FIG. In the controller according to the present embodiment , the adjustment coefficient C = 3 and the mode signal Q = 1, so that the internal model parameter determination unit 13 determines the gain Km of the internal model 6.
Is calculated as 3, the time constant Tm is 120 seconds, and the dead time Lm is 20 seconds, whereby the time constant T1 = 80.64 seconds and T2 =
20.16 seconds.

In the conventional controller shown in FIG. 12, the settling time when the stability adjustment is performed (y3) is 679 seconds, compared to the setting time 281 seconds when the filter stability adjustment is not performed (y1). It can be seen that responsiveness is impaired. On the other hand, in the controller (y4) of FIG. 1, the settling time is 777 seconds and the responsiveness is impaired.
In the controller (y) of this reference example , the settling time is 443 seconds, and the responsiveness is not impaired.

FIGS. 14 and 15 correspond to FIGS.
In the example of 13, the dead time Lp of the process to be controlled is 8
It is a figure which shows target value followability when it fluctuates to 0 second. In the conventional controller, the settling time when the filter stability is adjusted (y3) is 860 seconds, and the controller (y4) in the example of FIG. 1 has a settling time of 798 seconds, which is almost the same as the conventional controller. On the other hand, the controller (y) of this reference example
It can be seen that the settling time is 415 seconds and the stability is further improved, despite the fact that the identification error at the time of setting the process parameters has quadrupled with respect to the dead time Lp.

As described above, according to the controller of the present embodiment , the mode signal is set to the second mode, and the adjustment coefficient C for emphasizing stability is merely specified. Control can be realized without sacrificing control.

In the example of FIG. 1, the equations (10) to (1)
The internal model parameters were calculated using 2).
It is also possible and effective to cause the internal model parameter calculation unit 11 to perform the following calculation having the effect of the example of 0. Km = C × Kp (16) Tm = C ^1/2 × Tp (17) Lm = Lp (18)

Alternatively, Km = C × Kp (19) Tm = 0.5 × (C + 1) × Tp (20) Lm = Lp (21)

FIG. 16 is a block diagram of a controller showing another embodiment of the present invention. Reference numeral 20 denotes an interface circuit for interfacing with an operator (not shown) or a device on the side of a process to be controlled, 21 denotes a read-only memory (hereinafter referred to as a ROM) serving as a program storage area, 22
Is a random access memory (hereinafter referred to as a variable storage area)
A RAM for calculating internal model parameters and performing an IMC control operation in accordance with a program stored in the ROM 21, a reference numeral 24, an address bus, a reference numeral 25, a data bus, and a reference numeral 26, a control bus.

This embodiment is another configuration example for realizing the operation of the controller shown in FIG. 1 or 10. The interface circuit 20 receives the target value r, the process parameters Kp, Tp, Lp, the adjustment coefficient C, and the mode signal Q input by the operator, receives the control amount y from the control target process, and calculates by the CPU 23. Manipulated variable u
Is output to the process to be controlled.

The ROM 21 stores a program for realizing the operation of FIG. 3 or FIG.
U23 sequentially reads out and executes these programs via the address bus 24 and the data bus 25, and outputs the operation result to the R via the address bus 24 and the data bus 25.
Stored in the AM 22 or the interface circuit 20
Output to the outside.

Next, the operation of such a controller will be described with reference to the example of FIG. The CPU 23 adjusts the adjustment coefficient C and the parameter K transmitted from the interface circuit 20.
The internal model parameters Km based on p, Tp, Lp,
Tm and Lm are calculated as described above (steps 100 to 1).
02) and store them in the RAM 22 (step 1).
03). Next, a target value filter operation is performed on the target value r sent from the interface circuit 20 (step 10).
4, 105), a first subtraction process is performed on the operation result rf (step 106), and the operation amount u is calculated from the operation result e1.
Is calculated and sent to the interface circuit 20 (step 107).

Subsequently, the CPU 23 calculates a reference control amount ym from the operation amount u (step 108), and performs a second subtraction process based on the control amount y sent from the interface circuit 20 (step 110). And step 104
Operations 110 to 110 are repeated every control cycle until the controller stops (step 111). As described above, the operation of the example of FIG. 1 can be realized, and the example of FIG. 10 can be similarly realized.

[0069]

According to the present invention, since the internal model is calculated by the internal model parameter calculating section calculating the internal model parameters from the process parameters and the adjustment coefficients, the adjustment coefficients for emphasizing stability are designated. Thus, control with good settling can be realized without sacrificing responsiveness. As a result, stable control can always be ensured, and occurrence of troubles due to control can be prevented, so that the work load of an operator who does not have specialized control knowledge can be reduced. Also, by simply designating an adjustment coefficient emphasizing responsiveness, control with a faster response than a conventional IMC controller in which responsiveness is adjusted by a filter can be realized.
Therefore, the production efficiency of the process to be controlled can be increased, and effects such as reduction in production cost can be obtained.

Further, the internal model parameter calculation unit uses the value obtained by multiplying the gain and the time constant in the process parameter by a specific value of 1 or less as the gain and the time constant in the internal model parameter, so that the responsiveness is not sacrificed. It is possible to obtain an internal model parameter capable of realizing control with good settability, and the internal model parameter calculation unit multiplies by a specific value greater than 1 to provide a faster response than a conventional IMC controller that has performed responsiveness adjustment with a filter. It is possible to obtain internal model parameters that can realize control.

[0071]

[Brief description of the drawings]

FIG. 1 is a block diagram of a controller showing one embodiment of the present invention.

FIG. 2 is a block diagram of a control system using the controller of FIG. 1;

FIG. 3 is a flowchart for explaining the operation of the controller in FIG. 1;

FIG. 4 is a diagram showing target value followability of a conventional IMC controller.

FIG. 5 is a diagram showing target value followability of the controller of FIG. 1;

6 is a diagram showing a change in the operation amount of the controller of FIG. 1 and a conventional controller.

7 is a diagram showing a change in the operation amount of the controller of FIG. 1 and a conventional controller.

FIG. 8 is a diagram showing target value followability of a conventional IMC controller.

FIG. 9 is a diagram showing target value followability of the controller of FIG. 1;

FIG. 10 is a block diagram of a controller showing a reference example of the present invention.

FIG. 11 is a flowchart for explaining the operation of the controller in FIG. 10;

FIG. 12 is a diagram showing target value followability of a conventional IMC controller.

FIG. 13 is a diagram showing target value followability of the controller in FIGS. 1 and 10;

FIG. 14 is a diagram showing target value followability of a conventional IMC controller.

FIG. 15 is a diagram showing target value followability of the controller in FIGS. 1 and 10;

FIG. 16 is a block diagram of a controller showing another embodiment of the present invention.

FIG. 17 is a block diagram of a control system using a conventional IMC controller.

[Explanation of symbols]

2 target value filter unit, 3 first subtraction processing unit, 4 operation amount calculation unit, 6a internal model storage unit, 6b internal model output calculation unit, 8 second subtraction processing unit, 9 adjustment Coefficient input unit, 10 ... process parameter input unit, 11, 11a
... internal model parameter calculation unit, 12 ... mode switching unit,
13 ... internal model parameter determination unit.

Claims (2)

(57) [Claims] An operation amount to be output to a control target process is calculated from a control target value, and the control target process is regarded as a first-order lag.
Control is performed by calculating a reference control amount corresponding to the control amount of the controlled process, which is the control result, using an internal model expressed by a transfer function having a dead time element, and feeding back the difference between the control amount and the reference control amount. In a controller having an IMC structure, a transfer function outputs a target value of input control with a time delay characteristic, and a first subtraction process for subtracting a feedback amount from an output of the target value filter unit A target value / disturbance filter unit for outputting the output of the first subtraction processing unit with a time-delay characteristic of the transfer function, and the internal module excluding the dead time element based on the parameters of the internal model.
An operation amount calculation unit including an operation unit that calculates and outputs an operation amount from an output of the target value / disturbance filter unit by a characteristic of a reciprocal of a transfer function of Dell; an internal model storage unit that stores the internal model parameter; of the internal model on the basis of the internal model parameters
An internal model output operation unit for calculating a reference control amount from the manipulated variable according to a characteristic of a transfer function; and a reference control amount output from the internal model output operation unit from a control amount of a control target process to obtain the feedback amount. A second subtraction processing unit to be output; a process parameter obtained by model-identifying the process to be controlled; and an adjustment coefficient for designating a control characteristic that emphasizes responsiveness or stability. A controller configured to calculate and store the internal model parameter in an internal model storage unit. 2. The controller according to claim 1, wherein the transfer function of the internal model includes a gain and a time constant.
Wherein is intended to use the internal model parameter, the internal model parameter calculating unit, when the adjustment factor indicates readiness emphasis gain in the process parameters,
A value obtained by multiplying the time constant by a specific value of 1 or less is defined as a gain and a time constant in the internal model parameter. When the adjustment coefficient indicates importance on stability, a specific value greater than 1 is set for the gain and the time constant in the process parameter. A controller wherein the multiplied values are used as a gain and a time constant in internal model parameters .